GLM 5.3 · Coding

GLM 5.3 for coding: our test runs

GLM 5.3 ran our 5 coding prompts in llmwise, through the same pipeline your messages take. Here's every reply as it came, whether it passed, what it cost and what it counts as on Pro, next to the other models.

Start free See pricing

Based on 5 of our test runs on September 27, 2026, through OpenRouter with the app's own prompt and settings. Updated October 15, 2026.

Short answer

In our test runs on September 27, 2026, GLM 5.3 passed 5 of 5 of our coding prompts (both hard ones): tied 1st of 20 models. Each reply counted as 1 message on Pro, where it has 250 a month, and cost $0.0015 to run on average.

Prompt by prompt

Each of the 5 coding prompts, and how GLM 5.3's reply did.

GLM 5.3 on each coding prompt
Prompt	Result	Cost	Time
Turn a title into a URL slugEveryday	Passed: All 7 tests passed.	$0.0010	0.4 s
Parse a duration like “1h 30m”Everyday	Passed: All 10 tests passed.	$0.0014	1.1 s
Merge overlapping intervalsEveryday	Passed: All 7 tests passed.	$0.0008	0.8 s
Evaluate an arithmetic expression, no evalHard	Passed: All 15 tests passed.	$0.0017	12.5 s
Parse CSV with quoted fieldsHard	Passed: All 8 tests passed.	$0.0024	2.6 s

Every coding reply from GLM 5.3

Turn a title into a URL slug
An everyday prompt: “Turn a title into a URL slug”, in full.
Passed: All 7 tests passed.
```
```javascript
function slugify(title) {
  return title
    .normalize("NFD")
    .replace(/[\u0300-\u036f]/g, "")
    .toLowerCase()
    .replace(/[^a-z0-9]+/g, "-")
    .replace(/^-+|-+$/g, "");
}
```
```
498 tokens in, 74 out · 0.4 s · $0.0010 · 1 message on Pro · answered by z-ai/glm-5.3 via Modal · September 27, 2026

Parse a duration like “1h 30m”

An everyday prompt: “Parse a duration like “1h 30m””, in full.

Passed: All 10 tests passed.

```javascript
function parseDuration(text) {
  if (typeof text !== "string") return null;
  const match = text.trim().match(/^(?:(\d+)h)?\s*(?:(\d+)m)?\s*(?:(\d+)s)?$/);
  if (!match || match[0].trim() === "") return null;
  const h = match[1] ? parseInt(match[1], 10) * 3600 : 0;
  const m = match[2] ? parseInt(match[2], 10) * 60 : 0;
  const s = match[3] ? parseInt(match[3], 10) : 0;
  return h + m + s;
}
```

503 tokens in, 161 out (4 of them reasoning) · 1.1 s · $0.0014 · 1 message on Pro · answered by z-ai/glm-5.3 via Wafer · September 27, 2026

Merge overlapping intervals

An everyday prompt: “Merge overlapping intervals”, in full.

Passed: All 7 tests passed.

```javascript
function mergeIntervals(intervals) {
  const sorted = [...intervals].sort((a, b) => a[0] - b[0]);
  const result = [];
  for (const [start, end] of sorted) {
    const last = result[result.length - 1];
    if (last && start <= last[1]) {
      last[1] = Math.max(last[1], end);
    } else {
      result.push([start, end]);
    }
  }
  return result;
}
```

486 tokens in, 114 out (1 of them reasoning) · 0.8 s · $0.0008 · 1 message on Pro · answered by z-ai/glm-5.3 via Wafer · September 27, 2026

Evaluate an arithmetic expression, no eval

A hard prompt: “Evaluate an arithmetic expression, no eval”, in full.

Passed: All 15 tests passed.

```javascript
function evaluate(expression) {
  if (typeof expression !== 'string') throw new Error('Invalid expression');
  const src = expression;
  let pos = 0;

  function skipSpaces() {
    while (pos < src.length && /\s/.test(src[pos])) pos++;
  }

  function parseExpr() {
    let left = parseTerm();
    for (;;) {
      skipSpaces();
      if (src[pos] === '+') { pos++; left += parseTerm(); }
      else if (src[pos] === '-') { pos++; left -= parseTerm(); }
      else return left;
    }
  }

  function parseTerm() {
    let left = parseUnary();
    for (;;) {
      skipSpaces();
      if (src[pos] === '*') { pos++; left *= parseUnary(); }
      else if (src[pos] === '/') { pos++; left /= parseUnary(); }
      else return left;
    }
  }

  function parseUnary() {
    skipSpaces();
    if (src[pos] === '-') { pos++; return -parseUnary(); }
    return parsePower();
  }

  function parsePower() {
    const base = parsePrimary();
    skipSpaces();
    if (src[pos] === '^') {
      pos++;
      return Math.pow(base, parseUnary()); // right-assoc, exponent may be unary
    }
    return base;
  }

  function parsePrimary() {
    skipSpaces();
    if (src[pos] === '(') {
      pos++;
      const value = parseExpr();
      skipSpaces();
      if (src[pos] !== ')') throw new Error('Invalid expression: expected )');
      pos++;
      return value;
    }
    if (src[pos] === '-') return parseUnary(); // handled here too for safety
    const start = pos;
    while (pos < src.length && /[0-9.]/.test(src[pos])) pos++;
    if (pos === start) throw new Error('Invalid expression: expected number');
    const text = src.slice(start, pos);
    if (!/^\d+(\.\d+)?$|^\.\d+$/.test(text)) throw new Error('Invalid number: ' + text);
    return parseFloat(text);
  }

  const result = parseExpr();
  skipSpaces();
  if (pos !== src.length) throw new Error('Invalid expression: unexpected trailing input');
  return result;
}
```

562 tokens in, 594 out (85 of them reasoning) · 12.5 s · $0.0017 · 1 message on Pro · answered by z-ai/glm-5.3 via Reka · September 27, 2026

Parse CSV with quoted fields

A hard prompt: “Parse CSV with quoted fields”, in full.

Passed: All 8 tests passed.

```javascript
function parseCsv(text) {
  if (text === '') return [];
  const rows = [];
  let row = [];
  let field = '';
  let inQuotes = false;
  let i = 0;
  const n = text.length;

  while (i < n) {
    const c = text[i];

    if (inQuotes) {
      if (c === '"') {
        if (text[i + 1] === '"') {
          field += '"';
          i += 2;
        } else {
          inQuotes = false;
          i++;
        }
      } else {
        field += c;
        i++;
      }
      continue;
    }

    if (c === '"' && field === '' && (i === 0 || /[\n\r,]/.test(text[i - 1]))) {
      inQuotes = true;
      i++;
    } else if (c === ',') {
      row.push(field);
      field = '';
      i++;
    } else if (c === '\r' && text[i + 1] === '\n') {
      row.push(field);
      rows.push(row);
      row = [];
      field = '';
      i += 2;
    } else if (c === '\n' || c === '\r') {
      row.push(field);
      rows.push(row);
      row = [];
      field = '';
      i++;
    } else {
      field += c;
      i++;
    }
  }

  // Only emit a final row if there's a pending field/row (avoids trailing newline creating empty row)
  if (field !== '' || row.length > 0) {
    row.push(field);
    rows.push(row);
  }

  return rows;
}
```

Notes on behavior:
- A quote only starts a quoted field at the beginning of a field (start of input, after a comma, or after a line break), per your third rule; otherwise `"` is literal.
- `""` inside quotes becomes a single `"`.
- Embedded `\n`/`\r\n` inside quotes stays in the field; a trailing newline at the end of input doesn't create an empty final row; `''` returns `[]`.

518 tokens in, 463 out (7 of them reasoning) · 2.6 s · $0.0024 · 1 message on Pro · answered by z-ai/glm-5.3 via Wafer · September 27, 2026

How it compares on coding

GLM 5.3 beside the job's picks among every model. 18 models share the hard-problems pick, GLM 5.3 among them.

GLM 5.3 beside other models on coding
Model	Passed	On Pro	Cost per reply
GLM 5.3Our pick: hard problems (shared), best value	5 of 5	250 a month on Pro	$0.0015
GLM 5.3 FlashOur pick: hard problems (shared), everyday	5 of 5	60 a day on Pro	$0.0003

On Pro: Pro's count on each model. Cost: what OpenRouter charged us per reply, on average.

Every limit is published. Paid plans also have a monthly fair-use limit on AI cost: Plus $3.75, Pro $7.50, Max $20, Ultra $42, Studio $85. Using every message on your plan at typical sizes stays under it; very large messages and heavy research use it faster. Every limit, explained.

GLM 5.3 next to each model on coding

GLM 5.3 against each model on the same 5 coding prompts. One run each, through OpenRouter: a wait depends on the provider and the load that day, so a lead under 10% counts as close.

Against Claude Fable 5.1: GLM 5.3 answered 5.2× sooner (1.1 s to 5.5 s), cost 28.9× less, and passed 5 of 5 to its 5.
Against Claude Opus 5.5: GLM 5.3 answered 6.8× sooner (1.1 s to 7.2 s), cost 11.6× less, and passed 5 of 5 to its 5.
Against Claude Sonnet 5.5: GLM 5.3 answered 1.9× sooner (1.1 s to 2.0 s), cost 4.3× less, and passed 5 of 5 to its 5.
Against Claude Sonnet 5: GLM 5.3 answered 2.5× sooner (1.1 s to 2.7 s), cost 6.9× less, and passed 5 of 5 to its 5.
Against Claude Haiku 5.5: GLM 5.3 answered 3.7× sooner (1.1 s to 3.9 s), cost 3.2× more, and passed 5 of 5 to its 5.
Against Claude Haiku 4.5: GLM 5.3 answered 2.2× sooner (1.1 s to 2.3 s), cost 1.9× less, and passed 5 of 5 to its 3. Only GLM 5.3 passed Evaluate an arithmetic expression, no eval; Parse CSV with quoted fields.
Against GPT-6 Astra: GLM 5.3 answered 7.8× sooner (1.1 s to 8.2 s), cost 15.2× less, and passed 5 of 5 to its 5.
Against GPT-6.1 Sol: GLM 5.3 answered 4.5× sooner (1.1 s to 4.8 s), cost 1.6× less, and passed 5 of 5 to its 5.
Against GPT-6 Sol: GLM 5.3 answered 4.6× sooner (1.1 s to 4.8 s), cost 3.6× less, and passed 5 of 5 to its 5.
Against GPT-6 Luna: GLM 5.3 answered 4.3× sooner (1.1 s to 4.6 s), cost 5.1× more, and passed 5 of 5 to its 5.
Against Gemini 4 Argon: GLM 5.3 answered 3.3× sooner (1.1 s to 3.5 s), cost 1.7× less, and passed 5 of 5 to its 5.
Against Gemini 3.1 Pro (preview): GLM 5.3 answered 9.7× sooner (1.1 s to 10.3 s), cost 13.2× less, and passed 5 of 5 to its 5.
Against Gemini 3.8 Flash: GLM 5.3 answered 4.0× sooner (1.1 s to 4.2 s), cost 1.3× less, and passed 5 of 5 to its 5.
Against DeepSeek V4.1 Flash: GLM 5.3 answered 2.4× sooner (1.1 s to 2.5 s), cost 1.2× less, and passed 5 of 5 to its 5.
Against DeepSeek V4 Pro: GLM 5.3 answered 113.3× sooner (1.1 s to 119.6 s), cost 16.3× less, and passed 5 of 5 to its 5.
Against Grok 4.7: GLM 5.3 answered 29.9× sooner (1.1 s to 31.6 s), cost 16.7× less, and passed 5 of 5 to its 5.
Against Kimi K3: GLM 5.3 answered 3.9× sooner (1.1 s to 4.1 s), cost 4.3× less, and passed 5 of 5 to its 5.
Against GLM 5.3 Flash: GLM 5.3 answered 7.9× sooner (1.1 s to 8.4 s), cost 5.8× more, and passed 5 of 5 to its 5.
Against Mistral Large 4: GLM 5.3 answered 34.9× sooner (1.1 s to 36.8 s), cost 5.0× less, and passed 5 of 5 to its 4. Only GLM 5.3 passed Evaluate an arithmetic expression, no eval.

How these runs were done

Tests. Automatic. The function runs against the prompt's tests in a separate Node.js process with a time limit and no file, network or child-process access; it passes when every test passes.

How the runs were done, and every coding prompt.

GLM 5.3, and coding, elsewhere

Questions

Is GLM 5.3 good for coding?

In our test runs it passed 5 of 5 coding prompts, tied 1st of the 20 models in llmwise. Every reply is on this page, so you can judge them yourself.

How many of my messages does a coding reply from GLM 5.3 use?

1 message each on Pro, where it has 250 a month on Pro. The price of a message is fixed and shown before you send it, however long the reply.

How were these runs done?

The same way for every model: each prompt sent through llmwise's own pipeline, each reply checked the same way. The methods page has every prompt and how each is scored.

Claude, GPT, Gemini, DeepSeek, Grok, Kimi, GLM, and Mistral, in one chat.

See what a message costs before you send it. Free is 5 messages to try; sign in with an email link, no password or card.